Circadian rhythms have been extensively studied as a behavioral model because of their high degree of predictability and reproducibility within a species, and because several components and regulatory mechanisms are conserved from Drosophila to humans. Molecular dissection of this process has begun in mice using knockout and transgenic strategies. In general, information from these approaches can be maximized if expression or excision of a gene can be restricted to a desired tissue and controlled. These approaches have been successfully used to investigate learning and memory, but have not yet been applied to the regulation of circadian rhythms. A major reason for this is that a reliable strategy has not yet been developed that can specifically target expression of a desired gene to the retinorecipient cells of the suprachiasmatic nucleus (SCN), a primary circadian regulator in mammals. A targeting approach will be developed here to achieve relatively specific expression of a gene product in the retinorecipient neurons of SCN, i.e. the primary cells in which the circadian clock is reset in response to environmental signals such as light. To show the utility of the targeting strategy, CRE recombinase will be expressed in these cells. Tissue-specific CRE-mediated gene excision will be demonstrated using an established reporter system. Once validated, the CRE-expressing mice will be breed with existing foxed NMDA1 receptor mice to test the hypothesis that NMDA receptors in retinorecipient neurons are critically required for light-induced phase shifts. The same targeting system may be adapted in the future to study the putative involvement of other signaling proteins in light-induced resetting of the circadian clock, such as CAM kinase II, NOS, cGMP- and cAMP-dependent protein kinases, MAP kinases and CREB. Moreover, the same targeting system should have utility in investigating individual components of the clock, such as PER, CRY, BMAL and CLOCK proteins.